The overall goal of the proposed research as an improved understanding of multiple demographic and social responses to dramatic social change. The setting is Nang Rong, Thailand, a rural locale within a country that until recently experienced rapid economic expansion and modernization, but now is attempting to adjust to a major economic reversal. The contrast between economic conditions before and after the devaluation of the baht in July, 1997, in combination with prospective and retrospective data about migration, remittances, and land use, provides a unique opportunity to study responses at the individual, household, and community levels. The proposed research involves data collection and analysis. Data collection consists of a household survey, a follow-up of selected migrants, a land survey, a community profile, and the purchase of remotely sensed data. The land survey is the most innovative of these. Using a combination of recently created cadastral maps and "roving discussion groups," households will be linked spatially to agricultural plots. The proposed surveys will be undertaken in the year 2000 and when they are integrated with data collected and assembled in prior projects, the resulting data set will include four waves of survey data tracking of migrants to principal destinations, retrospective event history data for migrants and non-migrants, and information about social networks as well as spatial reference of households and agricultural plots, maps showing major landscape features, aerial photographs, and satellite data. The proposed analysis focus on three major questions: (1) What effect did the economic downturn have on migratory behavior, living arrangements, and remittance patterns of young adults? (2) What is the impact of return migration on the local rural economy, especially as it relates to crop pattern, mode of cultivation and intensity of production of marginal lands? (3) What is the relationship between network structure and position and migration? Answering these questions builds on unique characteristics of the data to be collected, but in no way exhausts their potential.